Tuberculosis treatment and resulting abnormal blood glucose: a scoping review of studies from 1981 - 2021

Figures & data

Figure 1. PRISMA flow diagram.

Table 1. Description of the included studies.

FIRST AUTHOR (YEAR/COUNTRY)	AIM OF THE STUDY	STUDY DESIGN	SAMPLE SIZE	STUDY POPULATION	SEX (% MALES)	SIMPLE ANALYSIS*	ADVANCED ANALYSIS**
Purohit, S D (1984/India)	To assess the effect of rifampicin therapy on glucose tolerance	Prospective cohort	57	TB patients	77	Yes	No
Singh, M.M (1984/India)	To determine the prevalence of impaired glucose tolerance in active pulmonary TB patients and to determine the effect of anti-tuberculous chemotherapy on the glucose tolerance curves	Prospective cohort	52	TB patients	65	Yes	No
Oluboyo, P.O (1990/Nigeria)	To determine the significance of glucose intolerance in TB	Case-control study	54	Pulmonary TB patients	63	Yes	No
Jawad, F (1995/Pakistan)	To unmask glucose intolerance in patients with active pulmonary TB and to assess the effect of treatment on its reversal	Cross-sectional and prospective cohort	106	Outpatient pulmonary TB patients	59	No	No
Basoglu, O.K (1999/Turkey)	To compare glucose tolerance test results of pulmonary TB patients with those of patients with community-acquired pneumonia	Case-control study	58	Pulmonary TB patients	78	Yes	No
Tarbasi, P (2014/Iran)	To examine HbA1c of new TB patients and relate to if they complete TB treatment	Prospective cohort	158	New TB patients	49	Yes	No
Akinlade, K.S (2016/Nigeria)	To assess changes in glycated haemoglobin levels in MDR-TB patients	Longitudinal study	21	MDR-TB patients	68	Yes	No
Boillat-Blanco, N (2016/Tanzania)	To examine the association of TB and its outcome with the presence and persistence of hyperglycaemia in Tanzania, using three different DM screening tests.	Case-control study	530	Pulmonary TB patients	58	Yes	Yes
Lin, Y (2017/China)	To understand if blood glucose levels were stable or fluctuated during TB treatment	Prospective cohort	270	TB patients	66	Yes	Yes
Moreira, J (2018/Brazil)	To assess the impact/prevalence of hyperglycaemia on TB outcomes, comparison of treatment outcomes & one-year mortality rate based on the glycaemic status of patients during treatment	Retrospective cohort	473	HIV-TB coinfected patients	69	Yes	Yes
Diarra, B^# (2019/Mali)	To determine the prevalence of DM in newly diagnosed TB patients	Cross-sectional and prospective cohort	201	TB patients	73	Yes	Yes
Krishnappa, D (2019/India)	To determine the presence of hyperglycaemia (DM & IGT) in TB patients and assess outcomes after successful treatment	Prospective longitudinal study	582	TB patients	56	Yes	No
Habib, S.S (2020/Pakistan)	To investigate the outcome of bidirectional TB-DM screening in the private sector	Cross-sectional	6312	All patients attending a private clinic	53	No	Yes
Kubjane, M (2020/South Africa)	To assess the association between hyperglycaemia and TB, at TB diagnosis, and after three months of TB treatment	Prospective cohort	850	Patients with respiratory symptoms	53	Yes	Yes

^#Patients without DM at baseline were followed-up at months two and five.

*Chi-square, t-test, ANOVA, Fisher’s Exact, correlation, Mann–Whitney test, Kruskal–Wallis test

**Linear/Logistic/Multilinear regressions/Cox regression

Table 2. Glucose changes before and during treatment.

FIRST AUTHOR(YEAR)	SAMPLE SIZE	TIME OF GLUCOSE ESTIMATION	BASELINE DM (%)	BASELINE IGT (%)	PARTICIPANTS AT FOLLOW-UP (N)	% DM AT FOLLOW-UP	% IGT AT FOLLOW-UP
Purohit, S D (1984)	57	0, 7, 30 days	-	10.0	0	0.0	0.0
Singh, M.M (1984)	52	0, 4, 8, 12 weeks	3.8	44.2	52	1.9	11.5
Oluboyo, P.O (1990)	54	0, 3, 3 months PT^#	5.6	37.0	53	1.9	9.3
Jawad, F (1995)	106	0, ET	19.8	29.2	23	21.7	21.7
Basoglu, O.K (1999)	58	0, 3 months	8.6	10.3	58	0	0
Tarbasi, P (2014)	158	0, 3 months	0	31.0	158	24.0	34
Akinlade, K.S*(2016)	21	0, 2, 4, 6 months PT	-	-	-	-
Boillat-Blanco, N (2016) **	530	0, ET	6.8	24.3	378	1.4	10
Lin, Y (2017)	232	0, 2, 6 months	0	7.3	232	0	3.0
Moreira, J (2018)	473	0, 3, 6 months, ET, PT	2.1	10.4	426	5.4	-
Diarra, B (2019)	201	0, 2, 5 months	5.5	3.0	190	0	-
Krishnappa, D (2019)	582	0, ET, PT	7.0	4	579	-	1.5
Habib, S.S (2020)	6312	0, 3 months	24	32	502	-	42.0
Kubjane, M (2020)	850	0, 3 months	11.9	46.9	276	9.3	21.5

*Two months follow-up showed a significant reduction in mean HbA1c. No values were indicated.

**Reported values are for 2-hCG (two-hour capillary glucose). ^#In those with abnormal results. ET – End of treatment; PT – Post-treatment.

Follow-up glucose at 30 days and two months was grouped with three months

Figure 2. Timelines* for glucose estimation in different studies.

*Will not sum up to 100% since studies used a combination of timelines.

Table 3. Summary of results from the different studies.

FIRST AUTHOR (YEAR/COUNTRY)	OVERALL SUMMARY AND TREATMENT OUTCOME
Purohit, S.D (1984/India)	The mean rise in glucose was lower on the 30^th day compared to the pre-treatment rise. The mean rise in glucose was higher in the rifampicin group at different intervals compared to the baseline.
Singh, M.M (1984/India)	54% (19/35) of those with cavities had an impaired glucose tolerance indicating an association between cavities and IGT. Only six out of 23 with initial impaired glucose tolerance continued to show impairment at 12 weeks.
Oluboyo, P.O (1990/Nigeria)*	Only 3.7% of patients remained abnormal three months after full treatment compared to baseline. The result suggests glucose intolerance during TB treatment caused by infection and is reversible. All patients had improved symptomatic and radiographic features at the end of treatment.
Jawad, F (1995/Pakistan)*	Glucose levels improved and returned to normal after TB treatment.
Basoglu, O.K (1999/Turkey)*	OGTT results returned to normal in both TB and pneumonia groups after treatment. Cases with abnormal OGTT were older than 40 years and more likely to be males.
Tarbasi, P (2014/Iran)*	95% had a successful treatment outcome. 24% developed DM, were older, had the highest level of FBG and had the highest prevalence of cavitary lung lesions.
Akinlade, K.S (2016/Nigeria)	HbA1c decreased at two months post-treatment compared to baseline. There were no changes in glucose levels at months four and six of treatment compared to baseline. No difference was seen in HbA1c levels based on HIV status.
Boillat-Blanco, N (2016/Tanzania)	DM or IGT at enrolment was significantly associated with adverse TB outcomes (i.e. loss to follow-up, treatment failure, or death).
Lin, Y (2017/China)	HIV positive status, DM, smoking cigarettes and presenting at a hospital rather than a clinic increased the odds of association with unstable FBG.
Moreira, J (2018/Brazil)*	75% successful treatment outcome. Hyperglycaemia was associated with an increased risk of mortality one year after TB treatment compared to euglycaemia (48.9% vs 7.9%).
Diarra, B (2019/Mali)*	No elevated blood sugar was seen at follow-up for the two periods. 64% with DM had a good TB treatment outcome and blood sugar levels had no impact on treatment outcome.
Krishnappa, D (2019/India)	Patients with hyperglycaemia (DM & IGT) were older. The blood sugar levels improved in all patients with DM following treatment of TB.
Habib, S.S (2020/Pakistan)	42% (213/502) with previous normal HbA1c had an increased HbA1c at three months follow-up while 58% (141/244) with previous elevated HbA1c dropped to the normal range at three months follow-up.
Kubjane, M (2020/South Africa)	2.6% (n = 10) with DM at enrolment reverted to normal at follow-up; and 22.5% (n = 105) of patients with IGR reverted to normal at follow-up.

*Studies with outcomes

Table 4. Summary outcomes in HIV-positive patients receiving TB treatment.

FIRST AUTHOR(YEAR/COUNTRY)	STUDY POPULATION	SAMPLE SIZE	% HIV POSITIVE	OUTCOME BASED ON HIV STATUS
Tarbasi, P (2014/Iran)	New TB patients	158	3.2	Not assessed.
Akinlade, K.S (2016/Nigeria)	MDR-TB patients	21	26.3	No difference in baseline HbA1c between HIV-positive and negative participants (p = 0.954).
Boillat-Blanco, N (2016/Tanzania)	Pulmonary TB patients	530	32	Using HbA1c, HIV-positive participants had lower odds of DM (p = 0.048). No difference between those receiving ART and those not receiving ART.
Lin, Y (2017/China)	TB Patients	270	3.3	Only 9/151 with known HIV status were HIV positive and 6/9 HIV positive patients had unstable FBG. HIV-positive participants had higher odds of unstable FBG after adjusting for confounding (p = 0.027).
Moreira, J (2018/Brazil)	HIV-TB coinfected patients	473	100	Successful treatment outcome – euglycaemic group (75%), hyperglycaemia group (28%), DM group (80%);Adverse event – euglycaemic group (25%), hyperglycaemia group (71%), DM group (20%);Death – euglycaemic group (7.5%), hyperglycaemia group (51%), DM group (10%);Lost to follow-up – euglycaemic group (17%), hyperglycaemia group (20%), DM group (10%).
Diarra, B (2019/Mali)	TB patients	201	6.5	No HIV-positive participant developed DM during TB treatment.
Kubjane, M (2020/South Africa)	Patients with respiratory symptoms	850	61	Significant positive association between DM and TB at baseline (OR, 2.4 [95% CI, 1.3–4.3]) and follow-up (OR, 3.3 [95% CI, 1.5–7.3]) regardless of HIV status.